Microsoft Research’s Post

Recipient of an ICML 2025 Outstanding Paper Award, CollabLLM improves how LLMs collaborate with users, including knowing when to ask questions and how to adapt tone and communication style to different situations. This approach helps move AI toward more user-centric and trustworthy systems. https://msft.it/6043Shf29

  • Image llustrates the overall training procedure of CollabLLM. For a given conversational input, the LLM and a user simulator are used to sample conversation continuations. The sampled conversations are then scored using a reward model that utilizes various multiturn-aware rewards, which are then in turn used to update parameters of the LLM.
Raju Pawar

Turning Words into Influence & Impact Every brand has a story. Every Entrepreneur, Visionary Thinker has a message. But not everyone knows how to put it into words that matter. That’s where I come in!

1d

Congrats! 🎉

Like
Reply
Ariel Cohen

Founder @ Scaled a Bootstrapped SaaS to $2.5M ARR | Gen AI Expert

1d

Sounds like a game changer for keeping AIs on their toes—finally, they might start asking the right questions instead of just regurgitating data.

Like
Reply
Lars Liden

Principal Research Software Engineering Manager @ Microsoft Research | Neural Networks, Deep Learning AI

1d

👏

Like
Reply
See more comments

To view or add a comment, sign in

Explore topics